2026 年 9 巻 2 号 p. 476-485
Introduction: Suicide remains a major public health issue in Japan, where the suicide mortality rate is high across all age groups. While existing national statistics provide limited insight into the individual backgrounds of those who die by suicide, detailed forensic data offer an opportunity to explore the characteristics associated with different methods of suicide.
Methods: We conducted a retrospective study using anonymized data extracted from police documents submitted to the Osaka Medical Examiner's Office for medicolegal investigation. Among 1,129 individuals who died by suicide in 2017 and 2019, 669 cases with complete data were analyzed. Latent class analysis (LCA) was used to classify individuals based on variables, including suicide method, age group, sex, occupation, psychiatric consultation history, suicide attempt history, and living arrangement.
Results: Latent class analysis identified three distinct classes. Class 1 (38.1%) consisted mainly of middle-aged unemployed females with a history of psychiatric consultation, living with others, and frequently jumping from heights. Class 2 (35.1%) was characterized by elderly unemployed males with no history of psychiatric consultation and no suicide attempt history, predominantly using hanging. Class 3 (26.8%) comprised younger employed males with no history of psychiatric consultation and no suicide attempt history, also showing a high proportion of hanging and jumping from heights, and poisoning.
Conclusions: The use of LCA revealed distinct subgroups of suicide deaths characterized by background factors and method choice. These findings may aid in identifying vulnerable populations and inform the development of targeted suicide prevention strategies in Japan. More broadly, our results highlight the value of combining medicolegal information with data-driven classification methods to better understand suicide in other settings.
